Single-Cell Multiome Sequencing: Doublet Cell Detection Using Cell Clustering, Cell Type Annotation, DoubletFinder, scDblFinder & ArchR

Abstract

Single-cell sequencing continues to power an in-depth understanding of cell-to-cell heterogeneity and drive research discoveries. However, it is prone to sequencing technology-related inaccuracies such as the presence of doublet cells, two or more cells captured together and sequenced as one, which affect the downstream analysis and interpretation of single-cell sequencing data. Previous research has focused on detecting doublets in single-cell RNA sequencing (scRNA-seq) and single-cell ATAC sequencing (scATAC-seq) datasets. There is a research gap related to doublet detection in single-cell multi-omics data. I propose an ensemble approach for detecting doublets in single-cell multiome sequencing data with scRNA-seq and scATAC-seq profiles. This ensemble includes cell clustering, cell type annotation, doublet detection tools for scRNA-seq data: DoubletFinder & scDblFinder, and ArchR designed for scATAC-seq data. This mechanism showcases how to exploit the multi-omic nature of single-cell multiome sequencing data to identify doublet clusters and doublet cells.